Opening Data from the Ground Up Seven Steps To Open State, City, or Provincial Data Successfully how-to guide 25 August 2015 JAKARTA
How-to Guide: Opening Data from the Ground Up Seven Steps to Open State, City, or Provincial Data Successfully Open data is a powerful tool for civil society groups. Used correctly, it can enhance transparency and participation by putting actionable information in the hands of citizens. But, all too often, governments are unwilling or unable to open up their data. This how-to-guide is intended for civil society groups who want to convince local governments to disclose government data in open formats so that they can use this for their advocacy or development work. This guide was developed based on the lessons from the project Linking Freedom of Information and Open Data in Banda Aceh. The Web Foundation s Open Data Lab Jakarta designs and tests approaches that harness the power of open data to address real problems. For more details about the project, please visit labs.webfoundation.org or contact us directly at info@labs.webfoundation.org. For open government data to make an impact, the data disclosed needs to be relevant to the needs, priorities, and interests of the public. Ideally, the process of opening government data should start from the bottom up. In environments where government data is not yet open by default, this means that citizens, or groups representing citizen interests, should inform government regarding the types of data that should initially be proactively disclosed or prioritised for release. Based on our experience from the Linking Freedom of Information and Open Data in Banda Aceh project, we created this guide to describe the steps we have used to facilitate this process. Opening Data from the Ground Up 3
How-to Guide: Opening Data from the Ground Up Step 1 Identify a practical starting point: an issue or problem that open data can help to solve. Start by identifying a problem or issue that people want to solve. This could be a specific problem (e.g. unattended garbage, a high drop-out rate in public schools) or a larger issue (e.g. unspent health budgets). Public opinion research, focus group discussions, media analysis, or expert interviews can all be helpful for scoping these issues. But in our experience, conducting sector-wide consultations with different groups will allow for a better grasp of issues or problems that really matter to people, and uncover which of these issues are most amenable to data-driven solutions. The key here is to ensure that the issue or problem is truly important to people and to make the problem identification process as participatory as possible. Step 2 Prioritise key information needs or primary data sets based on the problem identified. Assemble key stakeholders from civil society, media, academia, and the business community who are working in or are knowledgeable about the particular issue to be tackled. Use participatory methodology to identify the data that the local government should proactively disclose (e.g. issue-decision matrix, problem-data notecards) so that it can be used by stakeholders in finding solutions to address the problems or issues identified in Step 1. Consulting the right people with the sector expertise related to the problem will ensure that data needs are appropriately articulated. To do this, a stakeholder mapping exercise that identifies key stakeholders working on the problem or issue identified may be helpful. We also made a step-by-step illustrated version of this how-to guide for Opening Data from the Ground Up. Check it out on the next page. 4 Opening Data from the Ground Up
Opening Data from the Ground Up Seven Steps to Open State, City, or Provincial Data Successfully 1 Identify a practical starting point: an issue or problem that open data can help to solve. Staff Background 2 Prioritise key information needs or primary datasets based on the findings in #1. Student-Tecaher Ratio School Performance Facilities Curriculum Budget and Spending 3 Convince the government to open up the data and identify (possible) blockages in the process. 5 Build the capacity of users in accessing, analysing, and utilising the data. 4 Assist government in opening data. 6 Hold a public event that allows for healthy dialogue between the citizens and the government. 7 Conduct an evaluation of the whole process and plan for future activities. End Opening Data from the Ground Up 5
How-to Guide: Opening Data from the Ground Up Step 3 Convince the local government to open up the data and identify (possible) blockages in the process. Hold a workshop with those government agencies that gather, curate, and manage the prioritised data sets, sharing citizen concerns with them. Explain the potential benefits of opening up data clearly. If possible, involve a representative of another local government that has experience in opening up data to share what the benefits are. Generate commitment from the local government to open up the determined priority data and assist in identifying the difficulties to do so. Jointly craft a timed action plan and agree on responsibilities in the data publication process. To identify the government agencies that produce and store data sets related to the issue or problem earlier identified, a rapid data inventory research can be conducted by interviewing key government offices regarding the data that they possess. Find a legal or regulatory framework at the national level that can be used as a basis in discussions with local government around the importance of proactive disclosure. This can be a Freedom of Information law, a data sharing policy, or an open data framework. Step 4 Assist government in opening up the data. Based on the action plan developed in Step 3, assist the agency or agencies responsible in opening up the data sets through training and/or coaching. If your organisation does not have the expertise to do this, partner with institutions (e.g. civic tech groups, local university, training institutes) with expertise in open data. Training normally involves the transfer of knowledge from a trainer to a group of participants with the application of skills done at a later date, while coaching is a mode of skills building where a coach assists a learner in achieving a specific personal or professional goal. Our experience suggests that coaching is the most effective process to embed open data skills - especially when it is conducted in-house. The coaching process should cover data topics ranging from availability assessment and quality evaluation 6 Opening Data from the Ground Up
How-to Guide: Opening Data from the Ground Up to publication. The length of this process depends on the capacity of the specific unit of the agency you engage with. In our experience in Banda Aceh, working with an information management unit with adequate human and technical resources is most effective. Step 4 is completed once the prioritised data sets are published online. Conducting a capacity analysis of government personnel assigned with information management tasks is important in order to determine the type of support that must be provided. This will also identify the length of time needed to ensure that desired competencies in data disclosure are achieved. Step 5 Build the capacity of data users in accessing, analysing, and utilising the data. Bring together the group that you consulted in Step 2 and train them in the basics of open data use including accessing, scraping, cleaning, and analysing data sets. As in Step 4, if you do not have the capacity to deliver this training, partner with organisations that have the open data expertise. It may be useful, depending on the audience, to include sessions on basic statistics, data s value for decision making and advocacy, as well as data visualisation (i.e. transforming data into visuals like graphs and charts) and creation of data narratives (i.e. creating stories out of data available). The training should use the data that was disclosed by the government and based on our experience, each group should work on a project with an actual output: an analysis paper, an advocacy letter, a blog entry, a news article, a powerpoint presentation, or infographics that can be widely disseminated in both online and offline formats. Optional, but to inspire participants, this can be done as a competition with a modest prize given at the end. Capacity assessment of user organisations in dealing with data (e.g. scraping, analysis, publication) is important in order to ensure that the capacity building intervention fits their needs. Opening Data from the Ground Up 7
How-to Guide: Opening Data from the Ground Up Step 6 Hold a public event to present what users have produced to government officials and hold an open discussion regarding the issues identified. To allow for the findings or insights from the data to be discussed, schedule a public event which will act as a platform for the user groups to present their outputs. Provide sufficient space for the local government to respond and open the floor for discussion. This should be managed to ensure that discussion is objective, productive, and non-confrontational. If done properly and successfully, this also bridges the two groups for future collaborative activities. An atmosphere of openness and trust is important to allow conversation between user groups and government. In a situation, for example, where governments are not willing and open to listen to citizen needs, it will be difficult to achieve the objectives of this step. Engaging a professional facilitator to moderate the process is important to ensure that productive discussion among stakeholders will occur. Step 7 Conduct an evaluation of the whole process and plan for future initiatives. To learn from the process and use it to inform future initiatives, conduct short meetings with key stakeholders to reflect on the experience. It is essential to talk separately with government officials and data users to allow them to be honest about their experiences. Identify strengths and weaknesses of the approach and reflect on the key results of the process. Finally, it is important that stakeholders plan for future initiatives. With government stakeholders, it can be a plan of how to replicate the open data initiative in other sectors if it is found to have been successful (e.g. health, transport, public infrastructure). With data user groups, it can be a plan on how to use data on an ongoing basis to inform advocacy and development work, and also to reach out to 8 Opening Data from the Ground Up
How-to Guide: Opening Data from the Ground Up organisations focusing on the other sectors. Learning necessitates that stakeholders are honest in assessing their own performance, able to accept failure, and happy to celebrate success. For a more objective assessment of the project, an external evaluator may be hired to assist stakeholders in this process. We encourage you to test the approach outlined above so that we can build evidence of how the approach can work in different sectors and different contexts. If you have questions regarding this guide and its accompanying resources, email us at info@labs.webfoundation.org. If you want to learn more about the project, email us at info@labs.webfoundation.org If you want to look at other open data projects, see labs.webfoundation.org If you want to give it a try and implement the approach - with or without us - let s talk! Opening Data from the Ground Up 9
resources: Opening Data from the Ground Up Seven Steps to Open State, City, or Provincial Data Successfully You can find the resources we created, tested, and used for this project on our website at labs.webfoundation.org/resources. These include: A sample workshop design we used in prioritising data sets with user groups in Banda Aceh. A sample workshop design we used in engaging with local government officials to open up education data sets. A sample training module (in Bahasa Indonesia) we used to train government officials in Aceh on open data. The skills handout (in Bahasa Indonesia) which we used to train user groups in Banda Aceh. The workshop design which we used for training the user groups can be found here. Some external resources and toolkits we recommend when you re starting with open data are: The World Bank s Open Government Data Toolkit (also at toolkit.dev.zognet.net). Open Knowledge s Open Data Handbook (also at opendatahandbook.org). Locally, there is also Perludem s Modul Open Data Pemilu (in Bahasa Indonesia, also at rumahpemilu.org). We hope you find them as useful as we did! Feel free to download, use, and adapt them. Let us know if you have any suggestions for further resources appropriate for this ground up approach, would like our insights to co-develop and make changes to any of these toolkits to fit your own projects, or simply want to give us feedback on them. 10 Opening Data from the Ground Up
Open Data Lab Jakarta labs.webfoundation.org @ODLabJkt info@labs.webfoundation.org World Wide Web Foundation webfoundation.org @webfoundation contact@webfoundation.org JAKARTA 2015 World Wide Web Foundation. This work is licensed under a Creative Commons Attribution 4.0 International License.